Back to EveryPatent.com
| United States Patent |
5,127,991
|
|
Lal
, et al.
|
July 7, 1992
|
Uniformity of copper etching in the fabrication of multilayer printed
circuit boards
Abstract
This invention is a process for etching copper sheets on insulating boards
for use in fabricating multilayer printed circuit boards. The improvement
resides in adding to a standard copper etching solution certain
alkyltrimethyl ammonium chlorides with alkyl chain lengths ranging from 6
to 20 carbon atoms in amounts sufficient to yield intermediate kinetics
behavior. Of special interest are dodecyltrimethyl ammonium chloride,
hexadecyltrimethyl ammonium chloride and octadecyltrimethyl ammonium
chloride present in an amount of from 0.01 to 1.0 wt %. A silicon-based
antiforming agent may be added in amounts of from 0.05 to 2.0 wt. % to
prevent an unacceptably large amount of foam from occurring during
etching. With this formulation, the rate is mass transport controlled at
low rates and almost independent of mass transport at high rates due to
the inhibition of the surface reaction rate. As a result, non-uniformities
in etch rates due to variations in mass transport conditions in processing
equipment may be eliminated. This unique formulation could significantly
improve the yields presently obtained in manufacturing as well as allow
the processing of fine-line multilayer circuit boards.
| Inventors:
|
Lal; Sudarshan (Glen Rock, NJ);
Smith; Craig G. (Bridgewater, NJ)
|
| Assignee:
|
AT&T Bell Laboratories (Murray Hill, NJ)
|
| Appl. No.:
|
709050 |
| Filed:
|
June 3, 1991 |
| Current U.S. Class: |
216/63; 216/62; 252/79.4; 252/79.5 |
| Intern'l Class: |
B44C 001/22; C23F 001/00 |
| Field of Search: |
156/656,659.1,666,345,901,902,640,642
252/79.2,79.4,79.5,142,156
134/3
|
References Cited
Other References
Veniamin G. Levich, Physiochemical Hydrodynamics 1962, Prentice-Hall, Inc.,
pp. 93-102.
John S. Newman, Electrochemical Systems 1973, Prentice-Hall, Inc., pp.
307-310.
R. E. Markle, Processing and Economic Aspects of Etchant Regeneration,
Plating and Surface Finishing, Jan. 1983, pp. 58"62.
V. V. Isaev et al., Zasch, Met., vol. 13, No. 4, Jul. 1977, pp,. 444-445.
A. F. Bogenschuetz et al., Chemical Abstracts, vol. 90, No. 161138z, 1979.
A. Tanaka, Chemical Abstracts, vol. 105, No. 105-106957u, 1986.
|
Primary Examiner: Powell; William A.
Attorney, Agent or Firm: Alber; Oleg E.
Claims
We claim:
1. The method of etching patterns in copper sheets on insulating boards for
use in fabricating printing circuits boards, which comprises contacting a
resist patterned copper sheet on an insulating substrate with an aqueous
copper etchant solution for a period sufficient to etch out resist
unprotected regions of the copper sheet, in which said etchant solution
comprises an alkyltrimethyl ammonium chloride with alkyl chain lengths
ranging from 6 to 20, the amount of said chloride in the solution being
sufficient to yield an intermediate kinetics behavior.
2. The method of claim 1 in which said alkyltrimethyl ammonium chloride
comprises at least one chloride selected from the group consisting of
dodecyltrimethyl ammonium chloride, hexadecyltrimethyl ammonium chloride
and octadecyltrimethyl ammonium chloride.
3. The process of claim 2 in which said chloride is added in an amount of
from 0.01 to 1.0 wt. %.
4. The process of claim 2 in which said alkyltrimethyl ammonium chloride
comprises dodecyltrimethyl ammonium chloride.
5. The process of claim 4 in which said dodecyltrimethyl ammonium chloride
is added in an amount of from 0.01 to 1.0 wt. %.
6. The process of claim 5 in which said dodecyltrimethyl ammonium chloride
is added in an amount of 0.1 wt. %.
7. The process of claim 6 in which said solution contains a silicone-based
anti-foaming agent in an amount of 0.1 wt. %.
8. The process of claim 1 in which said copper etching solution contains a
silicone based anti-foaming agent.
9. The process of claim 8 in which said anti-foaming agent is present in an
amount of from 0.05 to 2.0 wt. %.
10. The process of claim 9 in which said solution contains a silicone-based
anti-foaming agent in an amount of 0.1 wt. %.
11. The process of claim 1 in which said copper etching solution is a
cupric chloride solution containing 150 to 203 g/l (20-27 oz/gal) of total
copper, 1.6 to 2.4N HCl, and being used at a temperature of from
115.degree. to 130.degree. F. (46.degree.-55.degree. C.).
12. The process of claim 1 in which said solution is an alkaline copper
etching solution containing 165 to 188 g/l (22-25 oz/gal) of total copper,
pH of 8.2 to 8.8, and being used at a temperature within a range of from
115.degree.-130.degree. F. (46.degree.-55.degree. C.).
Description
FIELD OF THE INVENTION
This invention is concerned with improved uniformity of patterns etched in
copper sheets in the process of fabricating multilayer printed circuit
boards.
BACKGROUND OF THE INVENTION
Copper etching is a crucial step in the processing of multilayer printed
circuit boards (MLPCB). Copper sheets, laminated to glass fiber reinforced
epoxy board substrates and destined to become inner layer circuitry, are
imaged and then etched typically using acid CuCl.sub.2 -based etchant
solutions to produce the inner layer circuitry on the substrate. Two or
more individual substrates are then bonded with inner layer circuitry
being inside of the MLPCB. An example of an intermediate stage of a MLPCB,
denoted as 1, is shown in FIG. 1. The intermediate stage comprises boards,
2, of a C stage glass fiber reinforced material, metal conductors, 3, and
bonding layers, 4, which may be of a prepreg or B stage material. The
intermediate stage may also include metallizations, 5, on the outer
surfaces of the boards, which eventually shall be patterned into the outer
circuitry of the MLPCB. Typically, alkaline copper etchants are being used
for this etching.
The acid cupric chloride etchant is used to produce about 80% of inner
layer boards, 85% of the print-and-etch boards, and 90% of the flexible
circuit boards manufactured in the United States. The presently used
etchant and associated equipment are adequate for relatively simple
circuits. However, significant non-uniformities in the etch rate across
the boards are observed, with variations in the etch rate across the board
ranging typically from 5 to 10% and in some instances up to 30%. This
suggests that the processing of these boards requires more stringent
control of the copper etching process. In addition, the multilayer boards
developed for computer applications are more complex and use smaller
conducting lines. Improving the etch uniformity should improve the fine
line etching capability.
Numerous attempts were made to improve the uniformity of the copper etching
process with CuCl.sub.2 etchant. For example, Isaev and coworkers indicate
that addition of metal chlorides increases the etching rate. See V. V.
Isaev et al., Zasch, Met., Vol. 13, No. 4, July 1977, pp. 444-445.
Improvement in the etching of metals using solutions containing both
ferric and cupric chloride by varying the pH, temperature, and salt
concentration has also been reported by A. F. Bogenschuetz et al.,
Chemical Abstracts, Vol. 90, No. 161138z, 1979. It has also been reported
that undercutting effects may be reduced by the addition of a mixture of
an anionic surfactant with a structure RO(CH.sub.2 CH.sub.2 .dbd.O).sub.n
SO.sub.3 R' and a nonionic surfactant of N-alkanolmonocarboxamide, where
R' represents either H or an alkali metal and R is a (C.sub.6-15) alkyl
group. See A. Tanaka, Chemical Abstracts, Vol. 105, No. 105-106957u, 1986.
However, while some progress was made in improving the etch rate, the need
for more stringent control of the etching process and more uniform etching
of copper is still present.
SUMMARY OF THE INVENTION
This invention is a process for etching copper sheets on insulating boards
for use in fabricating multilayer printed circuit boards. The improvement
resides in adding to a typical copper etching solution certain
alkyltrimethyl ammonium chlorides with alkyl chain lengths ranging from 6
to 20 carbon atoms in amounts sufficient to yield intermediate kinetics
behavior. Of special interest are dodecyltrimethyl ammonium chloride,
hexadecyltrimethyl ammonium chloride and octadecyltrimethyl ammonium
chloride present in an amount of from 0.01 to 1 wt %. A silicon-based
antifoaming agent may be added in amounts of from 0.05 to 2.0 wt. % to
prevent an unacceptably large amount of foam occurring during etching.
With this formulation, the rate is mass transport controlled at low rates
and almost independent of mass transport at high rates due to the
inhibition of the surface reaction rate. As a result, non-uniformities in
etch rates due to variations in mass transport conditions in processing
equipment may be eliminated. This unique formulation could significantly
improve the yields presently obtained in manufacturing as well as allow
the processing of fine-line multilayer as well a dual-sided circuit boards
.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a schematic representation of an intermediate stage in MLPCB
production, with inner metallization being inward of the MLPCB.
FIG. 2 is a representation of a variation of an etching rate with square
root of disk rotation speed for mass transport controlled kinetics as well
as for intermediate kinetics;
FIG. 3 is a schematic representation of a design of a rotating disk
assembly for use in testing mass transport characteristics of the etchant
solutions;
FIG. 4 is a diagram of a variation of etch rate with .omega..sup.1/2 for
the modified cupric chloride etch containing 0.12 and 0.22%
dodecyltrimethyl ammonium chloride (DTAC) plus 0.26% (by weight) Dow
Corning 544 (DC-544) antifoaming agent, 87.5 g/l Cu.sup.+2 as cupric
chloride, 3N HCl, 55.degree. C.; the upper solid curve marked KCl contains
additional 1.0M KCl;
FIG. 5 is a schematic representation of a configuration of spray
chamber/reservoir used for stability experiments;
FIG. 6 is a diagram of a stability of cupric chloride etchant containing 2M
HCl, 0.1% DTAC and 0.1% DC-544;
FIG. 7 is a diagram of stability of cupric chloride etchant nominal
solution containing 2M HCl, 1M NaCl, 0.1% DTAC and 0.1% DC-544.
DETAILED DESCRIPTION
This invention is a process and chemistry for improviing the uniformity of
etching process and, thus, of etched circuitry on copper-clad fiberglass
reinforced substrates in the process of fabricating printed circuit
boards. Applicants have realized that the non-uniformity of the etching
process may be attributed to certain characteristics of the processing
equipment. In operation, boards with copper sheets thereon pass
horizontally between two nozzle matrices which spray the etchant onto the
boards. On the top surface of the board, the solution puddles in the
middle of the board, while on the edges and the bottom surface of the
board, the solution is continuously swept off by the spray. Therefore, the
etch rate is smaller in the middle of the top surface of the board. These
observations imply that mass transport, the process by which reactants
reach the surface and products are removed from the surface by convective
diffusion, plays a significant role in the reaction kinetics and that the
uniformity of the etching process may be improved.
Theoretical analysis of mass transport conditions at the surface of a
rotating disk indicates that the rate of transport of solution species is
linearly proportional to the square root of the rotation speed, .omega..
See Veniamin G. Levich, Physiochemical Hydrodynamics 1962, Prentice-Hall,
Inc., pp. 93-102. This characteristic of the rotating disk has been
verified by numerous workers in the field. For example, see John S.
Newman, Electrochemical Systems 1973, Prentice-Hall, Inc., pp. 307-310.
For the case of intermediate kinetics, at low rotation speeds, the rate of
solution transport varies as .omega..sup.1/2, see V. G. Levich, supra.,
but eventually the surface reaction limits the observed reaction rate and
the reaction rate becomes independent of the rotation speed. This
generalized behavior is illustrated in FIG. 2. Applicants have concluded,
regarding a standard or conventional cupric chloride etchant, that: 1) the
etch rate is mass transport controlled over the range of conditions used,
2) the etch rate increases significantly with Cl.sup.- concentration, 3)
the etch rate is not limited by Cu.sup.+2 for concentrated solutions (
>87.5 g/l copper), 4) the etch rate decreases with Cu.sup.+ concentration,
but to a smaller extent than the increase due to chloride ion, and 5)
cations used in combination with Cl.sup.- affect the rate, with the rate
increasing in the order of NH.sub.4.sup.+ >K.sup.+ >Na.sup.+ >H.sup.+.
In view of this, the applicants have realized that the uniformity of
etching across the board cannot be improved by varying the concentrations
of the standard etchant. Since the etch rate for the standard CuCl.sub.2
etchant is limited by mass transport under all of the conditions
indicated, improvements in the uniformity of the mass transport conditions
for the processing equipment could lead to the improvements in the etching
uniformity; however, there are limits to improvements in the processing
equipment including the reluctance on the part of the user to modify an
existing etching line. Therefore, if modifications to the chemistry of the
CuCl.sub.2 etchant would yield the intermediate kinetics condition
indicated in FIG. 1, the cupric chloride etching process would be less
sensitive to the characteristics of the processing equipment.
Etch rate experiments were performed using 2 oz. (2.8 mils thick) copper
clad glass-fiber reinforced epoxy board substrates which corresponded to
the material used in the manufacture. To facilitate the use of controlled
mass transport conditions for a study of the standard cupric chloride
etchant, using this material, a rotating disk assembly, 10, shown in FIG.
3 was used in the tests. The assembly includes a rotator shaft, 11, with
an enlarged end surface, 12, on which is positioned a copper clad disk,
13, secured to the rotator shaft by an etch-resistant screw, 14. The
copper clad disk of 1.2 cm diameter has a 0.6 cm (0.24 in.) hole in the
center to mount the disk to the rotator shaft. The disk was clad with 2 oz
(2.8 mils thick) copper layer on 60 mil thick glass-fiber reinforced epoxy
board substrate. The rotation speed was varied within a range of from 100
to 2000 rpm using a Pine Instruments Analytical Rotator Model ASR2. A 250
ml volume of the etchant was prepared for each run and experiments were
performed on freshly cleaned copper disks in a thermostat controlled cell.
The copper surface was cleaned, prior to the etching, using an alkaline
cleaner of the following composition: sodium hydroxide (5 oz/gal), sodium
carbonate (10 oz/gal), trisodium phosphate (8 oz/gal), dodecyl benzene
sulfonate (0.1 oz/gal), and the cleaner bath temperature was maintained at
60.degree. C. The copper disks were soaked in the cleaner for 1 minute,
rinsed in warm tap water for 30 seconds, rinsed twice with deionized
water, dipped in 10% sulfuric acid for 1 minute, rinsed in deionized
water, dried with methanol-air, and weighed on an analytical balance to
determine the starting weight of the disk. The disks were then mounted on
the rotator shaft. and etched for varying intervals of time. The etched
disks were removed from the shaft, washed, dried, and weighed. The etch
rate was subsequently calculated using the weight loss.
Typical operating parameters for the standard cupric chloride etchant in
use on copper etching processing lines are: 1). 150 to 203 g/l (20-27
oz/gal) total copper, 2). 1.6 to 2.4N HCl, 3). 31.degree. to 35.degree.
Be, and 4). 115.degree.-130.degree. F. (46.degree.-55.4.degree. C.).
Regeneration of the etchant involves bubbling chlorine gas to maintain an
oxidation-reduction potential (ORP) value of 520 mV on Pt vs Ag/AgCl. The
specific gravity is maintained by a bleed and feed procedure, adding a
2.0N HCl solution to replace the removed etchant.
Etching experiments were conducted under conditions including 87.5 g/l Cu
as cupric chloride, temperature of the etchant of 55.degree. C., and 3 or
4N HCl. Several etchant properties were analyzed for each experiment. The
normality of H.sup.+ was determined by titration with a standard NaOH
solution to a pH of 2.3 using a Mettler Autotitrator Model DL25.
Titratable chloride (Cl.sup.-) was determined by titrating a known sample
volume, acidified with dilute nitric acid, with a 0.1N silver nitrate
solution to a first derivative end point. The concentration of cuprous ion
was determined by first oxidizing Cu.sup.+ with excess ferric sulfate. The
equivalent ferrous ions formed was titrated with a 0.1N Ce(IV) standard
using ferroin as an indicator. The density of the solution (degrees Baume)
was measured with a hydrometer and the oxidation-reduction potential (ORP)
measured using a reference Pt-Ag/AgCl couple.
Applicants have discovered that addition to the CuCl.sub.2 solution of
certain alkyltrimethyl ammonium chlorides in amounts of from 0.01 to 1.0
wt. %, with further addition in some instances of silicone-based
antifoaming agents in amounts of from 0.05 to 2.0 wt. % led to the
improvement of the mass transport. The alkyltrimethyl ammonium chlorides,
selected from those having alkyl chain lengths ranging from 6 to 20,
preferably from 10 to 18, carbon atoms, yield the desired intermediate
kinetics behavior and etch rate which is very uniform across the disk.
Significant foaming occurred with 0.1 wt. % dodecyltrimethyl ammonium
chloride (DTAC) at 2000 rpm; however, the use of 0.26 wt. % of Dow Corning
544 (DC-544) antifoaming agent, resolved this problem. Two additional
members of the homologous series represented by DTAC were
hexadecyltrimethyl ammonium chloride (HTAC) and octadecyltrimethyl
ammonium chloride (OTAC). Table I (See Appendix) illustrates the etch rate
at 2000 rpm for the modified etch containing DTAC, HTAC, and OTAC.
Potential solubility problems may exist for the use of OTAC. The reduction
of the surface reaction rate increases with chain length; therefore, such
monovalent alkaline metal chlorides as KCl and NaCl may be added to the
etchant to yield a higher etch rate.
The preferred additive for the modified etch chemistry is dodecyltrimethyl
ammonium chloride with a nominal concentration of 0.1 wt. %. When used in
combination with 0.1 wt. % DC-544 antifoaming agent, intermediate kinetics
are obtained in which the etch rate is independent of mass transport
conditions at approximately >1500 rpm. The decrease in etch rate (without
additives) at 2000 rpm is from 1.2 mil/min to 0.7 mil/min. KCl or NaCl may
be added to the etchant to yield a higher etch rate, still exhibiting
intermediate kinetics. FIG. 4 represents the etch rate versus
.omega..sup.1/2 with the addition of DTAC and DC-544 (lower solid line)
and also of KCl (upper solid line).
To investigate the stability of these additives, a pilot scale experiment
was conducted to age the etchant by etching copper sheets in a spray
chamber. The configuration of the apparatus, 20, used for stability tests
is shown in FIG. 5, wherein etchant, 21, is being sprayed from nozzles,
22, onto a 2 oz (2.8 mils thick) copper sheet, 23. Regeneration of the
etchant was performed using hydrogen peroxide-HCl additions, with a
bleed-and-feed operation. A recipe for regeneration of cupric chloride
etchants involves the addition of 40 ml of stabilized 35% H.sub.2 O.sub.2
or 46.7 ml of stabilized 30% H.sub.2 O.sub.2 and 94.5 ml of 12M HCl for
each ounce of copper etched. See R. E. Markle, Processing and Economic
Aspects of Etchant Regeneration, Plating and Surface Finishing, January
1983, pp. 59-62, which is incorporated herein by reference.
Two nominal compositions of the modified etchant were investigated for
stability. Both compositions contained 90-130 g/l Cu and 2N H.sup.+. The
first solution was diluted with a diluent containing 0.1 wt. % DTAC and
0.1 wt. % DC-544. The second solution composition contained 1M NaCl at
make-up, and was diluted with a solution containing 0.1 wt. % DTAC, 0.1
wt. % DC-544, and 1M NaCl. The etching experiments were conducted at
55.degree. C. Normal operation involved oxidation of cuprous ions after
every two hours of etching operation. After four hours of operation, the
etchant was diluted to yield a composition of 90 g/l Cu and 2N H.sup.+.
The normal operation was then repeated for a total of 90 hours,
corresponding to about 5 turnovers of the etching bath.
The rotating disk behavior shown in FIGS. 6 and 7 indicates that
intermediate kinetics were maintained for both solution compositions after
5 turnovers of operation. No degradation of performance has occurred. The
high etch rate for the initial solution in FIG. 6 is due to a higher-than
normal concentration of H.sup.+. The solution properties for both
compositions are given in Tables II and III (See Appendix).
For comparison purposes several other modifications to the bath chemistry
were investigated: 1): the use of metal chlorides at significant
concentrations, and 2): the use of other additives including surfactants
at small concentrations. To investigate the effect of metal chlorides on
the etch rate, standard conditions of 87.5 g/l Cu.sup.+2 as cupric
chloride, 55.degree. C., and 4N additional Cl.sup.- were used. Rotation
speeds of 100 and 2000 rpm were used to test for dependency on mass
transport conditions. Cations investigated were NH.sub.4.sup.+, K.sup.+,
Na.sup.+, Li.sup.+, Ca.sup.+2, Mg.sup.+2, Ba.sup.+2, Mn.sup.+2, La.sup.+3
and Al.sup.+3. Different cations significantly affect the etch rate, but
the rate remained mass transport limited. Monovalent cations have the
greatest influence on the rate with relative etch rates being
NH.sub.4.sup.+ >K.sup.+ >Na.sup.+ >Li.sup.+. Divalent cations yield etch
rates smaller than those obtained with monovalent cations, including
H.sup.+. The relative rates are Ca.sup.+2 >Mg.sup.+2 >Ba.sup.+2
>Mn.sup.+2. For the trivalent cations, La.sup.+3 yields a larger etch rate
than Al.sup.+3.
To investigate the effect of other additives on the intermediate kinetics
and on the etch rate, standard conditions of 87.5 g/l Cu.sup.+2 as cupric
chloride, 55.degree. C., and 3N HCl were used. Rotation speeds of 1500 and
2000 rpm were chosen as being characteristic of the plateau region for
intermediate kinetics (FIG. 2). Several organic ammonium chlorides and
hydrochlorides were investigated for their effects on the intermediate
kinetics. Tetrabutyl ammonium chloride gives intermediate kinetics without
the need for an antifoaming agent. However, this additive tends to break
down with use as the etch rate for successive experiments increases. The
etch is also uneven, giving spotted etching over the millimeter scale.
Tetrapentyl ammonium chloride, tetrahexyl ammonium chloride, and
tetraoctyl ammonium chloride are all insoluble and as a result do not
inhibit the etching reaction. Cetyl pyridinium chloride has a limited
solubility and gives a spotty, non-uniform etch. Dodecylamine
hydrochloride and octadecylamine hydrochloride are also insoluble.
It is to be understood that the above-described improvements and tests
described with reference to acid cupric chloride etchant are applicable to
the alkaline copper etchants as well. For example, the uniformity of
copper etching with an alkaline etchant having 150 to 203 g/l (20-27
oz/gal) of total copper, 18.degree. to 26.50.degree. Be (1.145-1.225
specific gravity) 5-5.8M HCl, pH=8.2-8.8, useable at a temperature of
115.degree.-130.degree. F. (46.degree.-55.degree. C.), may be improved by
the addition of the specified alkyltrimethyl ammonium chlorides and
silicone-based antifoaming agents, as needed.
TABLE I
______________________________________
APPENDIX
Useful Additives for the Cupric Chloride Etchant
(87.5 g/l Cu.sup.++, 55.degree. C., 3 N HCl, 2000 RPM)
H.sup.+
Cl.sup.-
ORP .rho. R R
Additive N N mV g/cc Be.degree.
mg/s mil/min
______________________________________
2.66 5.35 461 1.190
23.0 0.436
1.190
0.057% DB-31
2.54 5.54 483 1.190
23.4 0.405
1.105
0.1% Dodecyl-
2.49 5.55 519 1.190
23.4 0.265
0.723
trimethyl
ammonium
chloride
0.1% Hexade-
2.44 5.53 388 1.200
24.0 0.214
0.585
cyltrimethyl
ammonium
chloride
0.1% Octade-
2.50 5.66 0.135
0.368
cyltrimethyl
ammonium
chloride
______________________________________
Note; All three trimethyl ammonium chloride solutions also contain 0.1%
DC544 antifoaming agent.
The octadecyltrimethyl ammonium chloride is insoluble at room temp.
TABLE II
__________________________________________________________________________
Results of Modified Etch Stability Experiment
Time Total Cu
Cu.sup.+
Cl.sup.-
H.sup.+
.gamma.
.eta. .rho.
Etch Rate DTAC
hours
Turnover
g/l g/l
N N dynes/cm
centipoise
g/cc
mil/min (2000 rpm)
g/l
__________________________________________________________________________
0 0.0 91.3 1.6
5.85
2.72
32.4 0.942 26.0
0.97 0.94
52 1.16 117.2
1.6
6.55
1.80
33.8 0.954 28.6
0.80 0.34
60 1.96 114.2
2.4
5.69
1.81
31.9 0.954 28.4
0.77 0.49
70 2.75 125.9
1.6
6.70
2.04
31.2 0.989 29.6
0.87 0.51
74 3.40 129.6
3.2
6.33
1.95
34.0 0.973 29.1
0.86 0.36
90 5.10 132.1
3.0
6.83
1.85
32.0 1.020 30.8
0.84 0.50
__________________________________________________________________________
Diluent used2 N HCl + 0.1% DTAC + 0.1% DC544
TABLE III
__________________________________________________________________________
APPENDIX
Results of Modified Etch Stability Experiment
Time Total Cu
Cu.sup.+
Cl.sup.-
H.sup.+
.gamma.
.eta. .rho.
Etch Rate DTAC
hours
Turnover
g/l g/l
N N dynes/cm
centipoise
g/cc
mil/min (2000 rpm)
g/l
__________________________________________________________________________
8 1.11 111.7
7.9
7.27
1.64
34.4 1.016 30.2
0.98 0.25
16 2.22 130.8
0.6
7.27
1.55
32.0 1.092 32.2 0.31
24 3.32 106.1
1.1
6.92
1.71
30.4 1.018 30.3
0.95 0.46
32 4.45 108.0
0.3
6.76
1.50
31.0 1.013 29.5 0.54
38 5.18 99.9 1.4
6.71
1.78
30.5 0.989 28.6
0.93 0.61
__________________________________________________________________________
Diluent used2 N HCl + 1M NaCl + 0.1% DTAC + 0.1% DC544
Top